Introduction

In a recent volume which offers an ample and perspicacious critical panorama of contemporary thought on the theme of “literariness,” Costanzo Di Girolamo (1981) proclaims the failure of every effort thus far attempted to isolate the specifically “literary”:

a) “Connotation” is not the monopoly of literary texts, but appears, in practice, in every linguistic act; and one cannot oppose, on this basis, literary language to standard language. The notion of “écart” does not meet the need.

b) The “non-referentiality” of literature is not a criterion of literariness: “The referential function can be slight or close to zero degree; or it can be extremely elevated, without a text thereby losing its literary quality within a certain social context” (108).

c) The concept of “poetic function,” or of autoreflexivity of the message, introduced by Roman Jakobson (1960)—which in substance opposes the poetic to the referential function—is not valid, inasmuch as the literary phenomenon is in any case one of communication (“a message in function of itself would produce non-communication: the poetic function in the pure state is the inexpressible, a non-linguistic act”)(110).

d) Neither is the Jakobsonian notion of the “dominant poetic” (1973), intended to resolve the aporia, valid. What is the dominant—asks Di Girolamo—of the famous electoral formula “I like Ike” adduced by Jakobson (1960)? “In fact,” he adds, “with the exception of laboratory monsters, no examples are given of a linguistic function that is seen to be clearly dominant over others” (110).

e) Neither is the Einstellung, or “set,” of the message an exclusive trait of literature. It is a procedure “common to any type of discourse and to any linguistic function, and a problem that regards the rhetoric of communication, that is to say, the setting of the message with all available, adequate, and pertinent stratagems: rhythm. euphony, alliteration, grammatical parallelism, construction of words, word choice, and so on. These devices can vary in quantity, but it would be difficult to do without them entirely” (111-112).

f) The notion of literary text as “hypersign”—with which Maria Corti (1978c, 89—114) develops one of Umberto Eco’s suggestions (1976a,125—142) does not appear valid to Di Girolamo, since he retains that in substance it is still tied to the referential/non-referential opposition” (47).

g) Teun A. van Dijk’s hypothesis (1972b) that the literary system would be endowed with adjunctive rules (rhyme, alliteration, specific lexicon, etc.), does not lead to any useful result given that “nothing stops such rules from appearing . . . in the publicity message as well” (47).

h) Neither is the Marxist sociological theory of literature as statement of cultural contradictions sufficient;, though applicable to the nineteenth-century novel, it can be extended neither to other genres nor, above all, to other periods.

i) Then, the definition proposed by neorhetoric—that literature would be characterized by the “figurality” of language—is ingenuous, for, as we know, rhetorical figures are constantly used in standard language as well.

j) Finally, the concept of the literary work’s “autonomy” is to be rejected: “One wants here vigorously to deny that literature, as such, is without any referential function; or that such function, when present, occupies a necessarily peripheral and negligible place” (108).

Di Girolamo concludes that one cannot define a literary text on the basis of intrinsic qualities or properties, and that it is “always the public—contemporary and/or future—which decides whether a text is literary or not” (109). “A text is literary’ only if and when a public exists that is educated, willing, and competent to recognize it as such; the poetic function is not intrinsic to the text,” but rather the exclusive result of the public reception of a work (109–110). If, then, one must speak of a “specific” quality, it should be specific not to literature but to “literary production,” that is, to the fact that the work is produced by a writer competent in this kind of production, but received solely by consumers who—in this sense at least—represent an incompetent public.

Di Girolamo affirms, in substance, that what is called “literariness” is constituted by a set of pertinent traits chosen arbitrarily by the culture from the variety of a language’s possible expressive functions, that this set, just because it is arbitrary, is a variable, and that therefore not only is it vain to seek a quid that is constituted within and “exclusive” to literature, but further that the only possible research on the “specific” can be no more than a “history”—a variable, that is—of the conception of “literariness” in the evolution of socio-cultural systems.

There is no doubt that “literariness” is a convention, but a convention that—and Di Girolamo would certainly agree on this point—one must not see, surely, as something established only by the culture of its “consumers.” Original works always present themselves as “anomalous” in comparison with ruling socio-cultural systems; and on first impact, they even come to be considered—for just this reason—as not specifically literary. Thus they produce in the socio-culture the system of their “literariness,” and only later come to be classified as legitimate components of literary memory. In short, it is not the “social system,” but the work,that establishes “literariness.” This point substantially overturns the consequentially that one would have to deduce from Di Girolamo’s thesis. But aside from this, is it really true that in the convention “literature” some constants cannot be recognized, some characteristics, that is, that in the passage of time do not remain the same? To state the matter clearly, it will be necessary to distinguish between what comes to be established as “aesthetic”—and this constitutes a very strong, fundamental, and evaluative variable—and what comes to be established as “literary.” To my mind, the “literary” has an infinitely longer duration, and even within the indisputable historicity of all human phenomena, has such “constancy” that it can be considered a permanent trait.

This “constant,” it seems to me, can be recognized in a particular treatment of the communicative model, or, if one prefers, in an “anomaly” of the ordinary communication model. It consists in what we shall call the introjection of referents, which the patient reader will find discussed in the following pages; it is a kind of introjection that—as we shall see—does not fall back upon the concept of the absolute autonomy of the work, upon its non-referentiality.

This phenomenon—it should be stressed—can also be found in the form of a treatment of writing not originally conceived as literature. This does not invalidate the approach to the problem but on the contrary—at least to my mind—confirms its validity. On the basis of the introjective vision of a communication of the “historic document” or “scientific treatise” type, the former can be read as “story” or the latter as a “utopia” or fantastic “hypothesis,” just as a sacred text, once held to be the word of Truth as Divine message can later be read as the word of imagination. These would be cases that Di Girolamo would point to in support of his empirical-variabilistic thesis, but to my mind they show rather how it may be possible to postulate a constant convention of “literariness.”

In the pages that follow, I shall try to circumstantiate and make explicit these concepts and others connected to or implied by them; and the discourse will prove to be as long as required to review substantially the entire, and quite complex, phenomenon of literary communication in its rapport with natural linguistic communication.